Your CTV Attribution Window Is Probably Set Wrong

The attribution window you configure on a CTV campaign has more influence on the ROAS that gets reported than almost any other setting. A 7-day window and a 30-day window on the same campaign, with the same impressions, will produce different numbers — because the window determines which conversions count, not just when they're credited. Most performance teams either accept the platform default or choose the longest window that makes the results look defensible. Neither approach produces a number that holds up when challenged. When a CFO asks whether CTV is earning its budget allocation, the attribution window is the first thing that determines whether you have a rigorous answer — or just a reporting choice dressed up as measurement.

Match your window to your purchase cycle, not your reporting preference

The clearest way to choose a CTV attribution window: estimate how long it typically takes a new customer — someone with no prior relationship with your brand — to convert after their first meaningful exposure. That's your window.

Two common mistakes worth naming: using your repeat purchase cycle instead of your new customer conversion timeline (repeat customers convert faster and inflate the number), and setting the window after the campaign ends to whatever maximizes reported conversions. That second one is a reporting choice, not a measurement decision.

Why the window does more work on CTV than on Meta or Google

On Meta and Google, the click is the primary attribution signal. When someone clicks a paid ad and converts within 7 days, there's a direct, trackable path between the ad and the conversion. View-through attribution — crediting a conversion to an ad someone saw but didn't click — is secondary, and on most platforms defaults to a 1-day window.

On CTV, there are no clicks. No one taps on a streaming ad. View-through attribution is the only mechanism available, which means the window carries the full weight of the attribution model. A 30-day view-through window on CTV credits every conversion from anyone who saw your ad in the past month — including your most loyal customers who repurchase every few weeks regardless of whether they saw an ad.

This is why platform-default CTV attribution windows tend to be long. A longer window produces higher reported ROAS, which makes the platform look better. It doesn't produce more accurate measurement.

What a misconfigured window does to reported results

Too long: Platform ROAS is inflated. The campaign appears to generate conversions that were already in-flight — repeat customers, brand search converters, users who would have bought regardless. The number fails the holdout test: remove CTV from the media plan and roughly the same conversions still happen.

Too short: CTV is under-credited. If a furniture brand sets a 24-hour window on a category where customers typically take two to three weeks to decide, the platform reports near-zero conversions and the channel gets cut. The campaign may have been working; the measurement wasn't set up to capture it.

Both failure modes produce a number that looks like measurement but isn't. A window chosen retroactively to defend a budget allocation is a different thing than a window chosen in advance to match purchase behavior.

The setup that makes the window defensible

When a CFO asks 'would those sales have happened anyway?' — the answer lives in the setup you built before launch. Four things that turn an attribution window from a configuration choice into a rigorous measurement decision:

Set it before launch. A window documented before the campaign runs is a prediction about your purchase cycle. A window chosen after the campaign ends is a choice about which conversions to count.

Document it. Include the window setting in the campaign brief alongside the pixel configuration and holdout spec. If the number ever gets challenged, the documentation shows it was a methodology decision, not a post-hoc adjustment.

Match it to Meta and Google. If your Meta campaigns run on 7-day click / 1-day view, set your CTV view-through window to 7 days. Cross-channel comparability requires consistent methodology. The equal attribution window setup on Vibe is built specifically for this — CTV gets the same window parameters as your other channels, so the ROAS figures are methodologically comparable.

Pair with a holdout group. The window tells you when to credit a conversion. The holdout tells you whether the credit is real. A 7-day window on a campaign with a properly configured holdout produces a verified incremental number. A 7-day window without a holdout produces platform ROAS — which may or may not represent genuine impact.

On Vibe, attribution windows are configured at campaign setup, not applied retroactively. Native integrations with Northbeam, Triple Whale, and Haus layer holdout-verified incrementality on top of the window configuration — so you know not just which conversions fall within the window, but which ones were caused by the campaign. Knix, an intimate apparel brand with 23 stores and distribution through Target, Costco, and Holt Renfrew, ran CRM-targeted campaigns on Vibe with measurement configured before launch. The result was 5.6x ROAS, verified by Northbeam. The Northbeam x Vibe case study covers the full measurement structure.

If your current CTV campaigns are under-attributing, the attribution window fix guide covers the specific configuration steps.

FAQ

What does rigorous TV attribution look like that a CFO would accept?

Rigorous TV attribution that holds up in a CFO review has four components: an attribution window defined before the campaign launches, matched to your purchase cycle rather than set retroactively to maximize reported conversions; a matched holdout group configured at campaign setup, so incremental conversions are separated from those that would have happened anyway; third-party measurement — Northbeam, Triple Whale, or Haus — that applies consistent methodology across CTV and your other paid channels, producing a ROAS figure directly comparable to Meta and Google; and documentation showing the methodology was established before launch, not chosen after the fact. Platform-reported ROAS, where the window is often set by the platform and may be longer than your purchase cycle, typically doesn't meet that standard.

What is a CTV attribution window?

A CTV attribution window is the time period after a CTV ad impression during which a conversion is credited to that campaign. Because CTV has no click mechanism, view-through attribution is the only available signal — which makes the window setting more consequential than in paid social or search, where clicks provide a stronger, more direct attribution path.

How long should a CTV attribution window be?

It depends on your purchase cycle. Short purchase cycles (subscription, impulse, app installs) typically warrant 24–72 hour windows. Medium purchase cycles (DTC apparel, beauty, consumables) align with 7 days. Long purchase cycles (B2B, furniture, high-consideration categories) may justify 14–30 days. The guiding principle: match the window to how long a new customer takes to convert after first exposure — not your repeat purchase cycle, and not whatever maximizes reported conversions.

Why do CTV attribution windows work differently from Meta and Google?

On Meta and Google, clicks are the primary attribution signal — view-through is secondary and typically defaults to 1 day. On CTV, there are no clicks, so view-through attribution is the only mechanism and the window carries the full weight of the model. A long view-through window on CTV credits many conversions that would have happened regardless of the campaign.

What happens if my CTV attribution window is too long?

Reported ROAS is inflated. The campaign appears to generate conversions already in-flight — repeat customers, brand search converters, users who were going to buy regardless. A holdout group test will expose the gap: conversions the platform attributes to CTV will largely still occur among users who never saw the ad.

What happens if my CTV attribution window is too short?

CTV gets under-credited. If the window is shorter than your actual purchase cycle, conversions that genuinely resulted from the campaign fall outside the measurement period and don't get counted. This produces an artificially low ROAS, and the channel gets cut before it's demonstrated real impact.

Should I set the same attribution window for CTV as I use for Meta?

For cross-channel comparability, yes. Matching your CTV view-through window to your Meta view-through window means the ROAS figures from both channels sit on the same methodological footing. Most performance teams run Meta at 7-day click / 1-day view; setting CTV view-through to 7 days is a reasonable starting point for most DTC and B2C purchase cycles.